Update aggregate_spectra.R

[FedeMassetti]

I tried to address the problems presented by aggregate_spectra in handling multiple columns in the data of the SpectralDataFrame objects.
With this modification, as soon as all the non-numeric columns are listed as "id", everything runs smoothly (tested for mean and median on the australia dataset and on my own dataset).
The only problem is that the following message is given. But the output has always been alright so far.
Warning message:
In if (id %in% names(features(obj))) { :
the condition has length > 1 and only the first element will be used

This would be worth double checking since this is the first try I try programming in R!

[FedeMassetti]

There was some problem happening when using just one column as id. Now it should be fixed.
Except for the warning message I mentioned before there is just one bug remaining that I cannot fix. Plyr needs to be loaded after spectacles otherwise it cannot run properly giving the following error.

Error in ddply(s, id, colwise(fun), ...) :
could not find function "ddply"

The strange thing is that if I run s <- ddply(s, id, colwise(mean)) alone I get the same error with or without the updated function running. But as a part of the use of aggregate_spectra it doesn't give any problem until the updated function is loaded.
The error doesn't appear also if require(plyr) is placed at the beginning of the script.